The Utah Autism Program Project aims to identify genetic and non-genetic subgroups of autism by studying possible phenotypic manifestations of etiologic heterogeneity and familial transmission in autistic children and their first-degree family members. Control groups of families with probands affected with non-specific mental retardation or dyslexia are also under study. The investigators have found a high frequency of the MHC C4B null allele in children with autism and their mothers. For the twenty-six probands with completed typing, half have had the null allele. For the thirty mothers typed, 47% have the null allele. This compares with the rate of 20% found in normal controls. In seventeen autistic children with completed HLA typing, 47% have manifest the same extended haplotypes previously found to be increased when compared to controls. A third area of immune genetic function focuses on the hypervariable region of the DR(1 allele. Approximately half of the autistic children have specific alleles that are functionally identical. Other immunologic defects in the population studied have been low levels of IgA in plasma and an increased frequency of the B-cell antigen D8-17 in autistic children. Taken together, the results obtained to date strongly suggest that immune factors are associated with the development of autism. Cognitive studies have demonstrated that mean scores for executive function fall significantly below that predicted by full-scale IQ scores in children with autism. Other studies that are part of this protocol include volumetric measurements of the brain done by magnetic resonance imaging scans. Digitalized data have been collected and are now being processed.